I've used ChatGPT, Claude and Gemini since they launched. My chat history became a graveyard of good ideas, so I built the missing layer.
Hey IH đź‘‹
Quick context: I've been using ChatGPT, Claude, and Gemini
since the day they launched. Years of daily use. And
somewhere along the way I realized something painful.
Every great answer, every useful insight, every "I'll come
back to this" moment was just gone. Buried in an infinite
scroll with no search, no highlights, no notes.
So I built Foldif. It's a Chrome extension that adds a
knowledge layer on top of all three AIs.
🎥 https://www.youtube.com/watch?v=GUz-vehLTpw
Here's what it actually does:
→ Highlight any text in an AI response (5 colors, Kindle-style)
→ Attach margin notes to highlights. Your thinking, saved
with the answer
→ Auto-detect TODOs from AI responses in 13 languages,
drop them in a kanban
→ Move a conversation from ChatGPT to Claude in one click
(pick last 5/10/20 turns, it lands in the target composer)
→ Compile your highlights into a real PDF/HTML book with
the Book Builder
→ Unlimited nested folders, full-text search across every
chat (FlexSearch, all local)
Stuff I'm proud of:
- Everything is saved to the cloud so you can access your
highlights, notes, and folders from any device. Sync is
end-to-end encrypted (AES-256) with your own passphrase.
Even we can't read your data. - 13 languages with full RTL support. I'm Turkish so this
mattered to me personally. - No signup wall. Install, use. 5-day Pro trial activates
automatically, no credit card.
Pricing (looking for honest feedback here):
- Free: folders, pins, notes, basic search, 25 bookmarks,
10 prompts, 3 cross-AI handoffs lifetime - Pro: everything unlimited plus highlights, margin notes,
TODOs, Book Builder, encrypted cloud sync - Founder Lifetime: one-time, first 100 customers only
Two things I'm genuinely wrestling with and would love your
take on:
-
Is the free tier too generous? I want people to actually
use it before they pay, but I worry the line between
free and Pro isn't sharp enough. -
Cross-AI handoff is my favorite feature but most users
don't discover it for days. How would you surface a
"hidden gem" without an annoying onboarding tour?
Link in profile. Roast it, question it, ask anything.
Pricing, stack, why I made certain calls. I'll be here
all day.
The strongest feature here may not be highlights or folders by themselves. It is the cross-AI handoff.
Most people already treat ChatGPT, Claude, and Gemini like separate thinking rooms, but the context gets trapped inside each one. If Foldif can make useful ideas, notes, TODOs, and handoffs move across tools cleanly, that feels more like an AI workspace layer than just a bookmark extension.
That framing may also help with your pricing question. The free tier can stay useful, but Pro should probably own the “serious knowledge workflow” moments: cross-AI handoffs, encrypted sync, Book Builder, TODO extraction, and deeper retrieval. Those are the features that make someone feel they are not just saving text, they are building a personal operating layer over AI.
The only thing I’d watch is the brand. Foldif is interesting, but it may not immediately carry the bigger promise if this becomes the layer between all major AI tools. A cleaner workflow-style name like Xevoa.com would fit that broader direction well because the product is really about moving, organizing, and reusing AI thinking across tools.
The product solves a real problem. The key is making people understand quickly that this is not another clipper. It is the missing layer between their AI conversations and actual execution.
Really appreciate this — the "workspace layer between AI tools"
framing is sharper than how I've been describing it. I've been
leading with "highlight + organize" but you're right that the
handoff is what makes it different from a clipper.
Going to think hard about pulling cross-AI handoff, encrypted
sync, Book Builder, and TODO extraction up as the Pro anchors.
That also answers part of my pricing question — if Pro is
clearly "the serious knowledge workflow tier," the free/Pro
line stops feeling arbitrary.
On the "not another clipper" point — that's exactly the
positioning problem I keep hitting. Most people install
expecting a save button and don't discover the handoff for
days. Working on how to surface that earlier without an
intrusive tour.
Thanks for taking the time to write this out.
That makes sense. If users install expecting a save button, the product has to show the handoff value before they mentally file it under “clipper.”
I’d probably treat the first-use moment less like onboarding and more like a proof moment: save one useful AI answer, turn it into a TODO, then move it into another AI/tool as usable context.
That makes the workspace-layer idea obvious without needing a tour.
Drop your email and I’ll send over a tighter version. I can map the first-use flow, Pro/free split, and the positioning angle in a cleaner way than crowding the thread.